This renewal application addresses the questions raised by the results obtained in the present period. It focuses on the regression and progression phases that the DMBA induced skin tumors in rabbits present. In the results obtained so far we have found H-ras to be activated in benign and self-regressing tumors (keratoacanthomas, K.A.s) at a significantly lower frequency than in squamous cell carcinomas (SCCs). This is occurring in humans and rabbits. The mutant H-ras allele is expressed during the mature and regressing phases of the KA. We will now test the hypothesis that H-ras is involved in the KA regression by analyzing an array of in vitro and in vivo systems. We will use transient expression assays in rabbit corneal epithelial cells, stable expression in cell cultures and long term phenotypic experiments in transgenic rabbits injected with several ras constructions under the control of an inducible keratin promoter. This will be combined with the study of the TGF beta 1, 2, and 3 expression in this system, their possible induction by ras oncogenes and its correlation with differentiation in this system. Complementary, and based on our preliminary results, we will study other genes that could be involved in the process of progression to SCC. To that effect we will complete the cloning of a dominant oncogene detected in a rabbit SCC and we will analyze its molecular structure and role in skin tumors progression.. The molecular characterization of this unique system that displays both regression and progression should be very instrumental in dissecting the tumorigenesis process and the role of ras in it.